Much of the evaluation and tuning of a search system relies on relevance labels--- annotations that say whether a document is useful for a given search and searcher. Ideally …
Retrieval and ranking models are the backbone of many applications such as web search, open domain QA, or text-based recommender systems. The latency of neural ranking …
J Wang, JX Huang, X Tu, J Wang, AJ Huang… - ACM Computing …, 2024 - dl.acm.org
Recent years have witnessed a substantial increase in the use of deep learning to solve various natural language processing (NLP) problems. Early deep learning models were …
Neural information retrieval systems typically use a cascading pipeline, in which a first-stage model retrieves a candidate set of documents and one or more subsequent stages re-rank …
Pre-trained transformer models, such as BERT and T5, have shown to be highly effective at ad hoc passage and document ranking. Due to the inherent sequence length limits of these …
Query expansion aims to mitigate the mismatch between the language used in a query and in a document. However, query expansion methods can suffer from introducing non-relevant …
As information retrieval researchers, we not only develop algorithmic solutions to hard problems, but we also insist on a proper, multifaceted evaluation of ideas. The literature on …
Societal biases resonate in the retrieved contents of information retrieval (IR) systems, resulting in reinforcing existing stereotypes. Approaching this issue requires established …
Users of Web search engines reveal their information needs through queries and clicks, making click logs a useful asset for information retrieval. However, click logs have not been …